FINANCE DATA PROJECT

This project we will focus on exploratory data analysis of stock prices. We'll focus on bank stocks and see how they progressed throughout the financial crisis all the way to early 2020.

Import necessary packages

In this exploratory data analysis, we will use pandas, numpy, datetime, matplotlib, seaborn, and pandas_datareader modules. These modules are preinstalled using pip.

import pandas as pd
import numpy as np
import datetime
import matplotlib.pyplot as plt
import seaborn as sns
import yfinance as yf
%matplotlib inline
sns.set()

Get the stock data

Using the pandas_datareader module, we will get the information for the following banks:

• Bank of America
• CitiGroup
• Capital One
• JPMorgan Chase
• Wells Fargo
• Keycorp

start = datetime.datetime(2006,1,1)
end = datetime.date.today()

bank_stocks = yf.download("BAC C COF JPM WFC KEY", start=start, end=end, group_by='ticker')

[*********************100%***********************]  6 of 6 completed

bank_stocks.columns.names = ['Bank Ticker', 'Stock Info']

ticker = ['BAC', 'C', 'COF', 'JPM', 'KEY', 'WFC']

Take a look on the bank_stocks data.

bank_stocks.head()

Bank Ticker	KEY						C				...	JPM				WFC
Stock Info	Open	High	Low	Close	Adj Close	Volume	Open	High	Low	Close	...	Low	Close	Adj Close	Volume	Open	High	Low	Close	Adj Close	Volume
Date
2006-01-03	33.180000	33.680000	32.900002	33.669998	21.695471	1847100	490.000000	493.799988	481.100006	492.899994	...	39.299999	40.189999	26.663109	12838600	31.600000	31.975000	31.195000	31.900000	20.523809	11016400
2006-01-04	33.540001	33.950001	33.459999	33.880001	21.830795	1815500	488.600006	491.000000	483.500000	483.799988	...	39.419998	39.619999	26.509205	13491500	31.799999	31.820000	31.365000	31.530001	20.285755	10870000
2006-01-05	33.849998	34.090000	33.730000	34.040001	21.933891	2057000	484.399994	487.799988	484.000000	486.200012	...	39.500000	39.740002	26.589512	8109400	31.500000	31.555000	31.309999	31.495001	20.263241	10158000
2006-01-06	34.040001	34.080002	33.689999	34.049999	21.940325	1290000	488.799988	489.000000	482.000000	486.200012	...	39.549999	40.020000	26.776842	7966900	31.580000	31.775000	31.385000	31.680000	20.382265	8403800
2006-01-09	34.000000	34.150002	33.799999	34.090000	21.966105	997500	486.000000	487.399994	483.000000	483.899994	...	39.880001	40.669998	27.211746	16575200	31.674999	31.825001	31.555000	31.674999	20.379047	5619600

5 rows × 36 columns

Find the max close price for each bank's stock

Maximum closing price:

for tick in ticker:
    print(tick, bank_stocks[tick]['Close'].max().round(5))

BAC 54.9
C 564.09998
COF 166.28999
JPM 166.44
KEY 39.79
WFC 65.93

Minimum closing price:

bank_stocks.xs(key='Close', axis=1, level='Stock Info').min()

Bank Ticker
KEY     4.60
C      10.20
COF     8.31
BAC     3.14
JPM    15.90
WFC     8.12
dtype: float64

Create a new empty DataFrame called returns. This df will contain the returns for each bank's stock.

returns = pd.DataFrame()

for tick in ticker:
    returns[tick + ' Return'] = bank_stocks[tick]['Close'].pct_change()

Here is the brief overview of returns dataset.

returns.tail()

	BAC Return	C Return	COF Return	JPM Return	KEY Return	WFC Return
Date
2021-08-02	-0.010428	0.001627	-0.009895	-0.004019	-0.006104	-0.003265
2021-08-03	0.015543	0.016979	0.008307	0.011378	0.024053	0.022712
2021-08-04	-0.007263	-0.005226	-0.001487	-0.010792	-0.018991	-0.014521
2021-08-05	0.019598	0.017075	0.012408	0.012629	0.013754	0.018202
2021-08-06	0.028960	0.019802	0.016913	0.028404	0.021608	0.037880

Create a pairplot using seaborn of the returns dataframe.

sns.pairplot(returns[1:])

<seaborn.axisgrid.PairGrid at 0x1b20aa6fe80>

Using the returns dataframe, figure out on what dates each bank stock had the best and worst single day returns.

Best single day returns:

returns.idxmin()

BAC Return   2009-01-20
C Return     2009-02-27
COF Return   2009-04-20
JPM Return   2009-01-20
KEY Return   2008-09-29
WFC Return   2009-01-20
dtype: datetime64[ns]

Worst single day returns:

returns.idxmax()

BAC Return   2009-04-09
C Return     2008-11-24
COF Return   2009-04-09
JPM Return   2009-01-21
KEY Return   2008-10-14
WFC Return   2008-07-16
dtype: datetime64[ns]

Take a look at the standard deviation of the returns. Which stock we classify as the riskiest over the time period? Which we would classify as the riskiest for the year 2015?

Standard Deviation of the returns over the time period.

returns.std()

BAC Return    0.032037
C Return      0.033917
COF Return    0.029581
JPM Return    0.024846
KEY Return    0.032525
WFC Return    0.027333
dtype: float64

Standard Deviation of the returns for the year 2019.

returns.loc['2019-01-01':'2019-12-31'].std()

BAC Return    0.014566
C Return      0.015354
COF Return    0.015220
JPM Return    0.011869
KEY Return    0.016226
WFC Return    0.012450
dtype: float64

Stock return with highest deviation is considered the riskiest because the price of that particular stock has higher volatality compared to the other.

Create a distplot using seaborn of the 2015 for KeyGroup

sns.distplot(returns.loc['2015-01-01':'2015-12-31']['KEY Return'], color='green', bins=50 )
plt.title("KEY Return for the year 2015")

c:\users\nzrnm\appdata\local\programs\python\python39\lib\site-packages\seaborn\distributions.py:2557: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).
  warnings.warn(msg, FutureWarning)

Text(0.5, 1.0, 'KEY Return for the year 2015')

Create a distplot using seaborn of the 2008 for CitiGroup

sns.distplot(returns.loc['2008-01-01':'2008-12-31']['C Return'], color='red', bins=50 )
plt.title("C Return for the year 2008")

c:\users\nzrnm\appdata\local\programs\python\python39\lib\site-packages\seaborn\distributions.py:2557: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).
  warnings.warn(msg, FutureWarning)

Text(0.5, 1.0, 'C Return for the year 2008')

Imports

Now we will import plotly and cufflinks as additional visualization modules.

sns.set_style('whitegrid')

import plotly.express as px
import plotly.graph_objects as go

Create a line plot showing Close price for each bank.

The plot below shows the Closing price for each stock from 2006 to 2016. For the first graph, we use pandas built in plotting library while the second graph we use the plotly library.

for tick in ticker:
    bank_stocks.loc['2006-01-01':'2015-12-31'][tick]['Close'].plot(label=tick, figsize=(16,6))
plt.legend()

<matplotlib.legend.Legend at 0x1b21204aca0>

px.line(bank_stocks.loc['2006-01-01':'2015-12-31'].xs(key = 'Close', axis=1, level='Stock Info'), template="plotly_dark")

Moving Averages

Plot the 30 day average against the Close Price for BofA stock for the year 2008

bank_stocks['BAC']

Stock Info	Open	High	Low	Close	Adj Close	Volume
Date
2006-01-03	46.919998	47.180000	46.150002	47.080002	34.280807	16296700
2006-01-04	47.000000	47.240002	46.450001	46.580002	33.916729	17757900
2006-01-05	46.580002	46.830002	46.320000	46.639999	33.960411	14970700
2006-01-06	46.799999	46.910000	46.349998	46.570000	33.909451	12599800
2006-01-09	46.720001	46.970001	46.360001	46.599998	33.931290	15619400
...	...	...	...	...	...	...
2021-08-02	38.400002	38.950001	37.910000	37.959999	37.959999	49894000
2021-08-03	38.060001	38.630001	37.549999	38.549999	38.549999	39680600
2021-08-04	38.029999	38.770000	37.860001	38.270000	38.270000	42289800
2021-08-05	38.570000	39.060001	38.549999	39.020000	39.020000	34754500
2021-08-06	39.470001	40.410000	39.470001	40.150002	40.150002	58000800

3926 rows × 6 columns

plt.figure(figsize=(16,6))
bank_stocks['BAC']['Close'].loc['2008-01-01':'2008-12-31'].rolling(window=30).mean().plot(label='30 day Moving Average')
bank_stocks['BAC']['Close'].loc['2008-01-01':'2008-12-31'].plot(label='BAC Close')
plt.legend()

<matplotlib.legend.Legend at 0x1b212056ac0>

Create a heatmap of the correlation between the stock Close Price.

sns.heatmap(bank_stocks.xs(key='Close', axis=1, level='Stock Info').corr(), annot=True)

<AxesSubplot:xlabel='Bank Ticker', ylabel='Bank Ticker'>

sns.clustermap(bank_stocks.xs(key='Close', axis=1, level='Stock Info').corr(), annot=True)

<seaborn.matrix.ClusterGrid at 0x1b21457f520>

Create a candle plot of BofA stock from Jan 1st 2015 to Jan 1st 2016

BofA = bank_stocks.loc['2015-01-01':'2016-01-01']['BAC']

fig = go.Figure(data=[go.Candlestick(x=BofA.index,
                open=BofA['Open'],
                high=BofA['High'],
                low=BofA['Low'],
                close=BofA['Close'])])

fig.show()